1

EMPIRICAL EVIDENCE FOR FREUD’S THEORY OF PRIMARY

PROCESS MENTATION IN ACUTE PSYCHOSIS

submission date: May 10th

, 2011

2

ABSTRACT

Freud (1895/1966; 1900/1953; 1915/1957) has proposed that primary

process functioning is typical for acute psychosis. A non-verbal method, the

„Geocat‟ (Brakel, Kleinsorge, Snodgrass and Shevrin, 2000), measures primary

processes operationalised as attributional categorisation, which considers

exemplars as similar if particular features match, even if these components are

arranged in a quite different configuration. With the use of GeoCat we explored

primary process mentation in 127 psychiatric patients. Results show that (1) there

are substantially higher levels of attributional choices in our sample of psychiatric

patients, independently of diagnosis, than in a non-patient population; (2)

psychotic patients tend to have more attributional choices than non-psychotic

patient; patients with acute psychotic symptoms show more attributional

responses than patients without acute psychotic symptoms; (3) this increase of

attributional choices with the psychotic condition is independent of self-rated

anxiety or medication intake. We propose that, instead, this increase of

attributional levels in the acutely psychotic patients reflects a predominance of

primary processing which is specifically tied to the acutely psychotic condition, as

proposed by Freud.

keywords: psychosis, primary process, Freud, measurement, GeoCat

3

INTRODUCTION

The issue of what constitutes psychotic thinking has been a long-standing focus of

interest in psychopathology. Building on previous work (Brakel, 2009; Brakel,

Shevrin, & Villa, 2002; Brakel, 2004; Brakel & Shevrin, 2005; Vanheule,

Roelstraete, Geerardyn, Murphy, Bazan, & Brakel, 2011), this empirical study

proposes to test if there is predominance of primary process mentation in

psychosis as was proposed by Freud by applying a new method that has shown

promise in mapping primary and secondary processes, called the GeoCat, an

abbreviation of “Geometric Categorisaton” (Brakel, Kleinsorge, Snodgrass, &

Shevrin, 2000; see further) in a population of psychiatric patients.

Primary and secondary processes

In Freud‟s concept of the mental apparatus (Freud, 1895/1966), the

primary process is earlier both in ontogeny and phylogeny and reflects the

primary function of the nervous system: the flight from incoming excitations by

the shortest pathway possible by means of free flowing quantities which follow

directly connected or contiguous neural pathways. At the mental level, the

primary process denotes mechanisms of association that are characteristic of

unconscious mental life, including “(…) faulty reasoning, absurdity, indirect

representation, representation by the opposite” (Freud, 1905/1960, p. 88-89).

Rapaport (1951, pp. 395-398) proposes that “pre-logical” states are dominated by

the primary-process mechanisms of condensation, displacement, the toleration of

contradictions, symbolisations, substitutions and “pars pro toto”. Holt (1967, p.

4

354) also speaks about an association on the basis of “non-essential” features. The

overall outcome is a search for identity of perception (Freud, 1900/1953, p. 602),

implying identity of visual but also phonological characteristics: “The ideas which

transfer their intensities to each other stand in the loosest mutual relations. They

are linked by associations of a kind that is scorned by our normal thinking and

relegated to the use of jokes. In particular, we find associations based on

homonyms and verbal similarities treated as equal in value to the rest.” (Freud

1900/1953, p. 596). The secondary process reflects a more developed functioning

of the nervous system: the search for an adequate act as a response to the actual

situation of need of the organism by means of a refrained flow of quantities under

the inhibitory influence of the ego (Freud, 1895/1966). The secondary process

functions to inhibit and control primary process tendencies that follow the

pleasure principle: this process is “attuned to the efficient attainment of goals in

reality with the delayed gratification of impulses that is necessary” (Holt, 2009,

p.3). At the mental level, the secondary process refers to rational thinking and can

be found in our waking and conscious thinking ruled by the reality principle

(Freud, 1911/1958). It builds on the “thought identity” or the content of ideas

(Freud, 1900/1953, p. 602), and functions to make logically plausible connections

between ideas, while ignoring the intensity of the excitation related to to them.The

distinction between these two principles of mental functioning has proven a useful

tool for many authors after Freud (for a recent review of literature see Vanheule et

al., 2010).

Measuring primary and secondary process mentation

5

The first effort to find non-clinical confirmation of Freud‟s primary

process theory of dream formation was undertaken by Schrötter (1911/1951) who

used hypnosis to track symbolic transformations in dreams of prior hypnotic

suggestions (see also Roffenstein, 1924/1951; Nachmansohn, 1925/1951). Using

brief exposures of pictures, Pötzl (1917) found that parts of a stimulus presented

below the perception threshold appeared in the manifest content of the subsequent

dreams but that these fragments had undergone significant distortions closely

resembling the mechanisms of the primary process (see also Fisher, 1954,

1957).For the assessment of the primary process, Shevrin and Luborsky (1961)

introduced the rebus method – e.g. a rebus composed of e.g. the pictures of a pen

and a knee - directly inspired from Freud‟s Interpretation of dreams. Primary

process reading of this subliminally presented picture-puzzle leads to the

phonemic condensation of the sounds and therefore to the new word penny or its

related associations, appearing in dreams following the subliminal presentation.

Moreover, for the first time, brain markers for secondary and primary process

effects were demonstrated linking the study of primary process to neuroscience

(see Shevrin, 1973 for a review of these studies). A recent study by Villa, Shevrin,

Snodgrass, Bazan and Brakel (2006), based on the same theoretical principles, has

brought further evidence that unconsciously words are treated as sensorimotor

objects while consciously the same words are treated as counters in meaning.

However, not all studies dealing with primary process thinking used the

method of subliminal stimulation. Rapaport, Gill and Schafer (1945-46) and Holt

(Holt, 1956, 1977, 2002; Holt & Havel, 1960) independently developed methods

to identify primary process in Rorschach responses . Recently, Holt has published

6

an extensive manual for his “Primary Process System” (PRIPRO-system; Holt,

2007, 2009) which assesses both content and formal characteristics of primary

process but also examines whether manifestations of condensation, displacement,

symbolization, contradiction and distortion can be discerned in people‟s

thinking.Auld, Goldenberg and Weiss (1968) constructed a rating scale fort the

scoring of primary process thinking in dreams. Finally, Martindale and Dailey

(1996) developed a computerized scoring system (the Regressive Imagery

Dictionary) that can be applied in computerized lexical analyses of a variety of

text materials. It focuses primarily on content characteristics of primary process

However, these different methods are often complex and lengthy and they

are at least partially based on content analysis, implying an interpretation of the

materials produced by the subject, which often requires clinical skills and

(extensive) training. Finally, these different methods are based on linguistic

materials, limiting their use in children or for cross-cultural comparisons. The

present study uses a geometric categorization task, called GeoCat (Brakel et al.,

2000), which is an example of a formal, non-verbal index of primary and

secondary process mentation which can be administered independently of the

psychoanalytic clinical interpretation or training as well as independently of

language. Previous studies have confirmed the validity of the GeoCat for the

measurement of primary and secondary process mentation (Brakel et al., 2000;

Brakel et al., 2002; Brakel & Shevrin, 2005; Vanheule et al., 2011).

The Present Study

7

We wished to determine if using GeoCat method we would find a

predominance of primary process mentation in psychosis as was proposed by

Freud. Indeed, in the Freudian model of psychosis, psychodynamic processes

typical for the unconscious are present in conscious mental life: “As regards the

relation of the two psychical systems [the conscious and the unconscious], all

observers have been struck by the fact that in schizophrenia a great deal is

expressed as being conscious which in the transference neuroses can only be

shown to be present in the Ucs. by psycho-analysis.” (Freud, 1915/1957, p .197).

Fenichel (1945) agrees: « In non-psychotic persons, this mode of thinking is still

effective in the unconscious. Therefore the impression arises that in schizophrenia

“the unconscious has become conscious”.» and he adds: « Because the „primary

processi‟ and the archaic ways of thinking have come to the fore again,

schizophrenics are not estranged by these mechanisms any more. ». Freud

(1900/1953, p. 568) underscores that psychosis, in this respect, is similar to the

dream: the same regression, which in the dream leads to predominance of the

primary process, is also seen in psychosis where it can lead to hallucinatory

regression. In particular positive psychotic symptoms, such as hallucinations,

perceptual distortions and delusions, function along primary process principles

(Freud, 1895/1966, pp. 326-327; 1900/1953, p. 605). Finally, Freud indicates that

the verbal transformations, typical for the language of schizophrenics (see e.g.,

Robbins, 2002), also reflect primary process functioning: « In schizophrenia,

words are subject to the same process as that which makes dream images out of

dream thoughts, the one we have called the primary process They undergo

condensation, and by means of displacement transfer their cathexes to one another

8

in their entirety. The process may go so far that a single word, if it is especially

suitable on account of its numerous connections, takes over the representation of a

whole train of thought. » (Freud, 1915/1957, p. 186; see also, Bazan, 2006).

Previously and more recently, several authors have argued along the same lines

that language use in schizophrenics typically reflects a lesser functioning of the

secondary processes (e.g. Bazan, 2007b; Bazan & Van de Vijver, 2009a, b;

Bonnard, 1969; Roulot, 2004).

On the basis of Freud‟s propositions according to which psychotic

thinking can be understood as a form of primary process mentation, we collected

GeoCat responses in a residential psychotic population and sought to explore

three questions: 1) will hospitalized (residential) psychiatric patients show

significantly more attributional responses compared to the healthy adults of a

previous study with the GeoCat; 2) will hospitalized psychotic patients show

more attributional responses than non-psychotic psychiatric patients and 3) will

psychotic patients in an acute hallucinatory or delusional state show more

attributional choices than patients without acute psychotic symptoms.

METHOD

Participants

The research included 127 psychiatric patients, either psychotic (n=72) or

not psychotic (n=55). The participants were recruited from three different

institutions: Clinique de la Borde in Cour-Cheverny (n=25) in France, Psychiatric

9

Centers Dr. Guislain in Ghent (n=59) and St.-Amandus in Beernem (n= 43) in

Belgium. Research language was either French (Clinique de la Borde) or Dutch

(Psychiatric Centers Dr. Guislain and St.-Amandus). The ethics oversight

committees of the respective institutions approved the research; individual

participants gave informed consent for the research as well as for access to their

medical files. All data obtained were registered and stored anonymously.

There are no significant demographic differences between the two

populations. The mean age of the psychotic versus non-psychotic patients is 40.3

(± 1.41) versus 42.0 (± 1.77) years ( standard error of the mean, SEM). The

gender proportion is 81/19 versus 73/27 in the psychotic respectively the non-

psychotic population. Both the psychotic and the non-psychotic group include

substantially more men than women; this is due to the fact that the Psychiatric

Center Sint-Amandus is (historically) a psychiatric center for male patients only.

The mean number years of education is 11.8 ( .24) and 12.4 ( .29) years and

the mean length of hospitalisation is 15.3 ( 1.21) and 11.8 ( 1.75) years for the

psychotic respectively the non-psychotic patients (±SEM).

Though psychotic patients take more neuroleptics than non-psychotic

patients (97.2 versus 73.1%; ²(2)=17.011, p<.001), remarkably, a very

substantial portion of the non-psychotic patients also receive neuroleptic

medication in their use as so called “behaviour stabilisators”. Psychotic patients

also have more anxiolytics (53.5 vesus 32.7%; ²(2)=6.926, p=.031, two-tailed

test); intake of other medication was comparable in both patient groups.

10

Diagnoses

Information concerning the differential diagnosis was obtained from the

treating psychiatrist and psychologist, who both knew the patient for at least 6

months. The first author was the treating psychologist at the psychiatric centre

Sint-Amandus in Beernem (n=47); K.V.D. and L.D.C. were psychology interns in

the psychiatric Center Dr. Guislain in Ghent (n= 43) and in Clinique de la Borde

in Cour-Cheverny in France (n=25) respectively, at the time of the studyii.

Patients diagnosed as “psychotic” had one the following DSM-IV-R-diagnoses:

schizophrenia [disorganized type, 295.10 (n=8); catatonic type, 295.20 (n=3),

paranoid type, 295.30 (n=43); residual type, 295.60 (n=2); schizoaffective

disorder, 295.70 (n=3) and undifferentiated type, 295.90 (n=5)] and psychotic

disorders [bipolar I disorder, 296.54 (n=2); delusional disorder, 297.1 (n=4) and

psychotic disorder NOS, 298.9 (n=2)]. Patients diagnosed as “non-psychotic” had

one the following DSM-diagnoses: personality disorders [301 (n=17)], mood

disorders [296 (n=13)], alcohol intoxication [303.90 (n=13)], adjustment disorders

[309 (n=4)], autistic disorder [299.00 (n=2)], dysthymic disorder [300.4 (n=2)],

cannabis intoxication [292.89 (n=1)], amnestic disorder [294.0 (n=1)], obsessive

compulsive disorder [300.3 (n=1)] and pathological gambling [312.31 (n=1)].

Procedure and Stimuli

All participants were personally asked for their voluntary participation and

briefly informed about what the research program included. They were not

promised any kind of compensation, financial or other and were told that they

could withdraw at any time. After giving consent, the participant was invited into

11

a quiet room and shown a GeoCat form. This form consists of six squares

presenting in the bottom both attributional and relational variants of a top

“master” figure (Brakel et al. 2002, p. 486)iii

.

The theoretical basis of the GeoCat is that primary and secondary

processes are reflected in two distinct modes of mental organization which are

called “attributional” and “relational” cognition in cognitive psychology (see

Brakel et al., 2000). Briefly, in attributional thinking exemplars are regarded as

similar if particular features or attributes match, even if the configurational

disposition of these features are quite different. As attributional similarity thus

associates objects on the basis of common but inessential features, it is proposed

to reflect primary process mentation. Attributional thinking is contrasted with

“relational” thinking which is a mode of cognitive categorization that builds on

logical relationships between even very different features and takes the total

configuration of these components into account. From a Freudian point of view,

this type of cognitive process is based on the thought identity of given cues and

reflects secondary process mentation.

The written instruction was read by the researcher: “On this form there are

six squares. Each square contains one item at the top and two below. Look at the

top central figure; decide which of the two choices below is more similar to the

top central one. Circle your answer to each.”. To this the researcher added that no

item was either correct or incorrect, and that we were simply interested in the

participant‟s own choice. The dependent variable was the number of

attributional choices (ATTs) on the GeoCat: it varies between 0 and 6 (the

number of relational choices or RELs then is the mirror reverse and varies

12

between 6 and 0). Two ways of calculating were used: (1) the mean (and median)

number of ATTs and (2) the % of participants favouring ATTs.

After the GeoCat, the participants completed the State-version of the

Spielberg‟s State-Trait Anxiety Inventory (STAI, Spielberger, Gorsuch &

Lushene, 1970; Spielberger, 1979). In Belgium the Dutch translation (Van Der

Ploeg, Defares and Spielberger, 1980) and in France the French translation

(Spielberger, Gorsuch, Lushene, Vagg and Jacobs, 1993) were used.

RESULTS

Attributional Responses in Psychiatric versus Non-psychiatric Adult

Population

A Kolmogorov-Smirnov (KS) goodness of fit test revealed that the

distribution of ATTs is non-normal in our sample of psychiatric patients

(KSZ=2.496; p<.001). Inspection of the histogram shows that the distribution of

ATTs in the psychiatric population conforms to a J-curve rather than to a normal

distribution (Figure 1, left graph). The Wilcoxon Signed-Ranks (WSR) test

(WSRZ=-3.043, p=.002), and Sign (S) test (SZ=-2.626, p=.009) show that ATTs

predominate over RELs to a significant degree. The psychiatric patients are

compared with a normal non-psychiatric population of a previous study called

LifeCat in which the GeoCat was administered in exactly the same manner

(Brakel et al., 2002), In this study, the distribution was already also non-normal

but conformed to an inverse J-curve (KSZ=4.634; p<.001; see Figure 1, right

Met opmaak: Engels(Groot-Brittannië)

13

graph). The Wilcoxon Matched-Pairs test (Z=6.70, p<.001) and Sign test (Z=7.45,

p<.001) in the non-psychiatric participant group of comparable age had shown

that RELs were selected significantly more often than ATTs.

= = = = = insert Figure 1 about here = = = = =

The mean number of ATTs in the present study (called “AcuteCat”,

psychiatric patients; 3.66±0.21) is more than double the mean in the LifeCat study

(non-patient participants; 1.76±0.14) (p<.001, one-tailed; Mann-Whitney U or

MWU=9389.0; Wilcoxon W or W=41774.0; Z=-6,941). The differences between

the two samples are even more clear-cut for the mediansiv (4 and 0) and the modes

(6 and 0) in the psychiatric, respectively the non-psychiatric samples. These

results are summarised in Table 1.

= = = = = insert Table 1 about here = = = = =

There is a highly significant shift in number of ATTs in the psychiatric

population when compared to a non-psychiatric population of comparable age;

psychiatric patients show a complete inversion of their response pattern, with 60%

of the psychiatric patients, selecting a majority of ATTs while in the LifeCat

sample, 24% selected a majority of ATTs. This difference is highly significant

(p<0.001; ²=50.870; see Figure 1). The skewness of the number of ATTs in the

AcuteCat sample is -.448, indicating a long left tail, while the skewness in the

LifeCat sample is +.955, indicating a long right tail.

14

The level of self-rated anxiety in this psychiatric population, as measured

by the STAI, is high: the mean self-rated anxiety is 45.01.2. Though we have no

anxiety measures for the LifeCat study, the STAI-values of the present study are

substantially and significantly higher than normal values, as given by the Dutch

(36.4; p<.001) or French (35.73; p<.001) norms for populations of comparable

age. The level of anxiety did not correlate with the number of ATTs in the total

psychiatric population (Spearman ρ=.009; p=.923).

Attributional Responses in Psychotic versus Non-Psychotic Patients

The distribution of ATTs is non-normal in both the sample of psychotic

and of the non-psychotic patients (KSZ=2.060 and 1.561; p<.001 and =.015

respectively). The distribution of ATTs in the psychotic population conforms to a

J-curve, while the distribution in the non-psychotic sample conforms to a U-curve

(Figure 2, left and right graphs respectively). In the psychotic sample, ATTs

predominate over RELs to a significant degree (WSRZ=-3.363, p=.001 and SZ= -

2.889, p=.004), while in the non-psychotic sample there is no significant

predominance of ATTs over RELs or vice versa (WSRZ=-.800, p=.424 and SZ=-

.549, p=.583).

= = = = = insert Figure 2 about here = = = = =

15

The mean number of ATTs in the psychotic patients (3.97±.27) is higher

than in the non-psychotic patients (3.25±.34) but the difference is only marginally

significant (p=.06, one-tailed; MWU=1671.5; W=3211.5; Z=-1.555). Both

samples have different medians (5 and 4 in the psychotic and the non-psychotic

patients respectively). These results are summarised in Table 2.

= = = = = insert Table 2 about here = = = = =

The difference between the proportion of psychotic versus non-psychotic

patients showing predominantly respectively ATTs (65.3% and 52.7%

respectively) or RELs (30.6% versus 43.6%), though in the expected direction, is

only marginally significant (p=.065, one-tailed; ²=2.291; see Table 2). The

skewness of the number of ATTs in the psychotic sample is -.690, indicating a

long left tail, while the skewness in the non-psychotic sample is -.160, indicating

a more or less symmetrical distribution (see Figure 2).

The levels of self-rated anxiety does not differ between psychotic and non-

psychotic patients (44.71.5 and 45.5±2.0 respectively) nor does it correlate with

the number of ATTs in any of the two populations (Spearman ρ=.041 and .001;

p=.742 and .994 for psychotics and non-psychotics respectively).

Attributional Responses in Patients with and without Acute Psychotic

Symptoms.

16

In a second step, we verify if we obtain more clear-cut differences if we

compare patients with acute psychotic symptoms (part of the psychotic patients)

and patients without acute psychotic symptoms (the rest of the psychotic and all

non-psychotic patients). Patients are judged to have acute psychotic symptoms if

at least one of the following positive symptoms was reported: hallucinations,

perceptual distortions, voices, intrusive experiences or delusions, at the moment

of or on the day preceding the experiment. Inpatients considered without acute

psychotic symptoms were those who registered no symptoms for at least a week

before and at least a weak after the experiment. Monitoring of positive symptoms

in the patients is continuous as patients are taken care of day and night by

competent nursing professionals (well staffed with approximately one nurse for

four patients); this staff meets five times a day, including three times a day with

the treating psychologist and once a week with the psychiatrist. The psychologist

has his or her consulting room in the residential department and/or is among the

patients when not consulting. By doing so the psychotic patient group sorted into

in two even groups: psychotic patients with acute psychotic symptoms (n=36) and

psychotic patients without acute psychotic symptoms (n=36)v.

The distribution of ATTs is non-normal in both the patients with and

without acute psychotic symptoms (KSZ=1.872 and 1.855; p=.002 and .002

respectively). The distribution of ATTs in the patients with acute psychotic

symptoms conforms to a J-curve, while the distribution in patients without acute

psychotic symptoms conforms to a U-curve (Figure 3, left and right graphs

respectively). In the sample with acute psychotic symptoms, ATTs predominate

over RELs to a significant degree (WSRZ=-3.275, p=.001 and SZ=-2.572,

17

p=.010), while in sample without acute psychotic symptoms while in the non-

psychotic sample there is no significant predominance of ATTs over RELs

(WSRZ=-1.522, p=.128 and SZ=-1.386, p=.166).

The mean number of ATTs in the patients with acute psychotic symptoms

(4.31±.37) is significantly higher than in the patients without acute psychotic

symptoms (3.41±.26) (p=.019, one-tailed; MWU=1262.0; W=5448.0; Z=-2.083).

Both samples have different medians (5 and 4 in the patients with, respectively

without, acute psychotic symptoms). These results are summarised in Table 3.

= = = = = insert Table 3 about here = = = = =

The proportions of patients with versus without acute psychotic symptoms

showing predominantly ATTs is 69.4% and 56.1% respectively. The difference is

more marked for the proportions of patients with versus without acute psychotic

symptoms showing predominantly RELs (25.0% versus 40.7%). These

differences are marginally significant (p=.056, one-tailed; ²=2.533; see Table 3).

The skewness of the number of ATTs in the sample with acute psychotic

symptoms is -.906, indicating a long left tail, while the skewness in the sample

without acute psychotic symptoms is -.294, indicating a more symmetrical

distribution (see Figure 3).

= = = = = insert Figure 3 about here = = = = =

18

The levels of self-rated anxiety, though lower in the group with acute

psychotic symptoms, did not differ significantly from the level in the group

without acute psychotic symptoms (42.42.3 and 46.0±1.5 respectively; t=-1.337;

p=.184). This level of anxiety did not correlate with the number of ATTs

(Spearman ρ=.032 and .017; p=.858 and .879 for patients with and without acute

psychotic symptoms respectively).

It thus seems that the psychotic patients without acute psychotic symptoms

behave as non-psychotic patients when it comes to their choices on the GeoCat

test. Indeed, the mean number of ATTs in these two groups (non-symptomatic

psychotics, n=36 and non-psychotics, n=55) did not differ (3.64±.39 and 3.25±.34

respectively) (p=.29, one-tailed; MWU=925.0; W=2465.0; Z=-.542).

The Influence of Medication on Attributional Choices.

As indicated, psychotic patients take significantly more neuroleptics and

anxiolytics than non-psychotic patients. When comparing patients with and

without acute psychotic symptoms, we found that, similarly, the patient group

with acute psychotic symptoms takes more neuroleptics (97.2 versus 82.8%;

²(2)=4.707, p=.030) and more anxiolytics (58.3 versus 39.1%; ²(2)=3.818,

p=.051) than the patient group without these acute symptoms (two-tailed tests).

However, the number of ATTs in the whole patient group did not correlate with

medication intake, neither for the neuroleptics nor for the anxiolytics (Spearman

ρ=-.104 and -.132; p=.250 and .146 respectively). Therefore, we have no

indications that the differences in medication between the groups could explain

19

the differences in ATTs. Note that there is also no correlation between the levels

of self-rated anxiety and the intake of anxiolytics or neuroleptics (r=.053 and -

.019; p=.567 and .838 respectively).

The potential interaction effect of the neuroleptics intake with the patient

group could not be verified, as there were only two psychotic patients,

respectively one patient with acute psychotic symptoms, not taking neuroleptics.

However, there were no differences between the two patient groups in

correlations between ATTs and neuroleptics intake (ρ=-.178; p=.138 versus ρ=-

.150; p=.281 in the psychotic versus in the non-psychotic patients and ρ=-.150;

p=.383 versus ρ=-.140; p=.195 in patients with and without acute psychotic

symptoms).We verified the potential interaction effects of the anxiolytics with the

patient group on the number of ATTs with a nonparametric method, the adjusted

rank transform test (Leys & Schumann, 2010). However, no interaction effect was

found, neither for the comparison between psychotic and non-psychotic patients

(Factorial Anova on the adjusted ranks, F=.260; p=.611), nor for the comparison

of the patients with and without acute psychotic symptoms (F=2.231; p=.138).

When subtracting possible interaction effects of the anxiolytics intake, the main

effect of the patient group on ATTs remained non-significant in the case of the

comparison between psychotics and non-psychotics (F=2.032; p=.157) and

significant in the case of the comparison between patients with and without acute

psychotic symptoms (F=5.438; p=.021) (two-tailed tests).

DISCUSSION AND CONCLUSIONS

20

Three questions were investigated with the GeoCat method in which the number

of ATTs are proposed to probe for primary process mentation: 1) will psychiatric

patients show significantly more ATTs compared to non-patients; 2) will

psychotic patients show more ATTs than non-psychotic psychiatric patients and

3) will psychotic patients in an acute hallucinatory or delusional state show more

ATTs than patients without acute psychotic symptoms.

First, in the present study it is found that, regardless of the psychotic

condition, the mean number of ATTs in the psychiatric patients is more than

double the mean in a non-patient adult population, (“LifeCat”) with about 60%

versus about 30% of ATTs respectively. Accordingly, it is proposed that

mentation is much more dominated by primary processes in residential psychiatric

subjects. This is not surprising as it may be assumed that general levels of anxiety

are substantially higher in this population, which correlates well with the self-

rated anxiety measure showing significantly above normal levels of anxiety in the

psychiatric population. Using the same GeoCat instrument, we have shown before

that anxiety correlates highly with primary process mentation (Brakel & Shevrin,

2005). In this respect it is important to note that in the present study the anxiety

measure did not correlate with the number of ATTs, However, it must be noted

that the levels are generally very high in this psychiatric population so that a

ceiling effect probably masks the sort of correlation previously found. Another

factor may be important for the attributional shift in the psychiatric population,

namely the infantilising effect expected from living in a residential setting. This

regressive influence is also thought to lead to more primary process mentation.

21

Therefore, we believe that the substantial shift to more ATTs correlates with a

genuine higher level of primary process mentation in a population of residential

psychiatric patients.

When comparing the results on the GeoCat, we found that psychotic patients had

a higher mean number of ATTs than non-psychotic patients, but the result is only

marginally significant. When we compare the patients with acute psychotic

symptoms with patients without acute psychotic symptoms , we do find a

significant increase. Therefore, we think that the increase of ATTs in the group of

psychotic patients is real but not large enough to become significant in the present

sample, and that it is carried by the subgroup of psychotic patients with acute

psychotic symptoms. Comparison of the distribution graphs in the different

populations indicates that the sensitive part related to the acutely psychotic

condition is the relative absence of subjects who give an “all relational” response

pattern, with zero attributional choices. This fits well with the proposition that

“relational” choices are based on secondary processes, and with stabilised

psychotic patients being indistinguishable from non-psychotic psychiatric patients

in that respect. In other words, it is psychotic decompensation which is

specifically related to a a relative absence of secondary processes and to primary

process predominance, producing both positive psychotic symptoms and

preferential attributional categorisation.

Finally, the influence of medication should be considered since there is

more neuroleptics and anxiolytics intake in the psychotic patients. The intake of

these medications does not correlate with the number of ATTs. This is not

surprising, since there was also no influence of the medication on the level of self-

22

rated anxiety. For comparison, we report two previous studies on neuroleptics,

where the effects of the typical antipsychotic phenothiazines on primary process

thinking were measured on schizophrenic patients. Research data were obtained

from pre- and postdrug Rorschach protocols, which were scored using Holt's

(1959) method for measuring primary process. In one study, as patients improved,

so did the control of primary process mentation (Saretsky, 1966); in the other

(Ebert, Ewing, Rogers, & Reynolds, 1977), there were progressive decreases on

formal primary process scores. In these studies, however, the within-subjects

design measured changes in each subject‟s primary process-parameters before and

after drug treatment, while in the present research, between-subjects measures are

employed which are not sensitive to the ameliorative effects of medications on

individual participants.

There are some limitations to this study, which are tied to the diagnostic

procedures. Differential diagnosis between psychotic and non-psychotic patients

is based upon the psychiatric diagnosis made by the treating psychiatrist on the

basis of the DSM-IV in accord with the treating (psychodynamically trained)

psychologist. Distinction between patients with and without acute psychotic

symptoms is based upon reports by the caregivers as described above.

Standardized questionnaires or checklist procedures were not used. Though this is

a limitation, we want to underscore that all the patients of this study were

residential patients for many years in the institution with well documented

medical files and that different authors of the study (A.B., K.V.D., L.D) were

practicing clinicians in the institutions at the time of the research.

23

In conclusion, our results show that there is an increased level of

attributional choices in psychotic patients when they are in an acute state of their

condition, and that that level is significantly higher than the already increased

levels in the other psychiatric patients without acute psychotic symptoms. As self-

rated anxiety levels are identical in the psychotic and non-psychotic population,

anxiety can not be related to this further increase. Instead, as suggested above, we

propose that this further increase of attributional levels in the acutely psychotic

patients reflects an emergence of primary process thinking, and a lesser

proportion of secondary processing, reflective of the acutely psychotic condition

as described long ago by Freud (1895; 1900; 1915) and others (Bleuler, 1911).

This also coheres with Holt (2002)‟s general conclusion, resulting from a review

of fourteen empirical studies based on Holt‟s PRIPRO-system in Rorschach

protocols in schizophrenia. Although the results are somewhat scattered, Holt

(2002: 474) concludes that, on the whole, they support the psychoanalytic

expectation that schizophrenia is “accompanied by the disruptive emergence of

primary process thinking into conscious thought”. These results are also in line

with the concept of “thought disorder” (e.g. Kasanin, 1944; Andreasen, 2008) in

schizophrenia, i.e. thinking which contains “incoherence, tangentiality, or

derailment (loose associations)” (Andreasen, 2008, p. 436). Remarkably Von

Domarus (1944) calls it “paralogical thinking”, namely thinking for which

identity is based on partial identification rather than on total identity.

The fact that we were able to show more attributional mentation in acute

psychosis confers supplementary convergent validation for GeoCat as a test of

primary process mentation. Obviously, this supports the test as a useful diagnostic

24

tool which may contribute valuable clinical information, indicating if a patient is

in a predominantly primary process mode at the moment of the testing. However,

some caution is advised since this new instrument is still in development.

Moreover, the predominance of primary processes is not indicative of psychosis

per se, as a diverse number of pathological conditions (such as trauma, anxiety)

and non-pathological states (such as regression, creativity, hypnosis) are also

characterised by primary process predominance. However, provided this caution,

the tool promises to be clinically most useful as it is rapid, non-linguistic, easy to

use and independent of clinical interpretation.

Finally, the present results also fit well with a number of recent

neuroscientific accounts for the Freudian primary and secondary processes, which

all associate the psychotic condition with a lesser control of secondary over

primary processes. Recently, indeed, the neuroscientists Carhart-Harris and

Friston (2010) have proposed that Freud‟s descriptions of primary and secondary

processes are consistent with self-organized activity in hierarchical cortical

systems where the secondary process provides top-down predictions to reduce

free-energy associated with the primary process. The authors propose that the

high-levels of this inferential hierarchy form a cortical network of regions which

they call the “default-mode network” (DMN). The DMN is a cortical network of

strongly interconnected nodes (including the medial prefrontal cortex, the

posterior cingulate cortex, the inferior parietal lobule, the lateral and inferior

temporal cortex and the medial temporal lobes) which show high metabolic

activity and blood flow at rest but which deactivate during goal-directed cognition

(Raichle, MacLeod, Snyder, Gusnard, & Shulman, 2001). Recent work has shown

25

reduced task-evoked suppressions of DMN activity in schizophrenia (Pomarol-

Clotet, Salvador, Sarró, Gomar, Vila, Martínez, et al., 2008; Whitfield-Gabrieli,

Thermenos, Milanovic, Tsuang, Faraone, McCarley, et al., 2009) the severity of

which correlated positively with connectivity in the DMN (Whitfield-Gabrieli et

al., 2009). Therefore, Carhart-Harris and Friston (2010) propose that

schizophrenia is associated with a loss of top-down control of the DMN over

limbic activity in hierarchically lower systems and that this is equivalent to a loss

of control from the ego and the associated secondary process over the primary

process (p. 3). This, of course, would fit well with the results of the present study.

We have proposed a parallel account for the Freudian primary and

secondary processes before (Bazan, 2007a, 2007b), which could also fit in

Carhart-Harris and Friston‟s view. Freud (1895, p. 326-327) had said: “[Wishful

discharges] to the point of hallucination and complete generation of unpleasure

which involves a complete expenditure of defence are described by us as

psychical primary processes; by contrast, those processes which are only made

possible by a good engagement of the ego, and which represent a moderation in

the foregoing, and are described as psychical secondary processes. It will be seen

that the necessary preconditions of the latter is a correct employment of the

indications of reality, which is only possible when there is inhibition by the ego.”.

The differentiating criterion between primary and secondary process in a Freudian

account, then, are the so-called “indications of reality”. Based on historical,

anatomical, functional and semantic arguments, a parallel between these

“indications of reality”, which Freud derived from a Helmholtzian model of

perception, and the recent “efference copy models”, which is also rooted in the

26

model developed by von Helmholtz, was proposed (Bazan, 2007a, 2007b).

Starting from this parallel, it was further proposed that the dorsal pathway

(involving dorsal and prefrontal cortices), which makes use of these efference

copies for the organisation of contextualised action, might be considered as a

neurophysiological correlate for the secondary process. This dorsal pathway

hierarchically controls selection of activations in the ventral pathway (Friedman-

Hill, Robertson, Desimone & Ungerleider, 2003; Hamker, 2003; Rousselet,

Thorpe & Fabre-Thorpe, 2004), which for this (and other) reason(s) might be

considered as a neurophysiological correlate for the primary process (Bazan,

2007a, 2007b). Indeed, selective impairment of the dorsal “where” route (but not

of the ventral “what” pathway) has been associated with the psychotic condition

(Daniel, Mores, Carite, Boyer, & Denis, 2006). Therefore, this dorsal/ventral

neurophysiological account also fits well with the diminished secondary processes

and increased primary processes in the acute stage of psychosis observed in our

data.

The results of this research then contribute to support three important

propositions: Freud‟s proposition that psychosis is characterised by a

predominance of primary process mentation, the usefulness of the Geocat as a

measure of primary process mentation, and the possibility of empirically testing

psychoanalytic hypotheses with methods independent of the psychoanalytic

clinic.

27

REFERENCES

Andreasen, N.C. (2008). Thought disorder. In S. H. Fatemi and P. J. Clayton

(eds.) The Medical Basis of Psychiatry. Humana Press, Part III, pp. 435-

443.

Auld, F., Goldenberg, G., & Weiss, J. (1968). Measurement or primary process

thinking in dream reports. Journal of Personality and Social Psychology,

8(4, Part 1), 418-426.

Bazan, A. (2006). Primary process language. Neuro-Psychoanalysis, 8, 157-159.

Bazan, A. (2007a). An attempt towards an integrative comparison of

psychoanalytical and sensorimotor control theories of action. In P.

Haggard, Y. Rossetti & M. Kawato (Eds.), Attention and Performance

XXII, pp. 319-338. Oxford University Press, New York.

Bazan, A. (2007b). Des fantômes dans la voix. Une hypothèse neuro-

psychanalytique sur la structure de l’inconscient. Editions Liber,

Montréal.

Bazan, A., & Van de Vijver, G. (2009a). L‟objet d‟une science neuro-

psychanalytique. Questions épistémologiques et mise à l‟épreuve. In: L.

Ouss, B. Golse, N. Georgieff, D. Widlöcher (eds.), Vers une

neuropsychanalyse? Odile Jacob, Paris, 33-54.

Bazan, A., & Van de Vijver, G. (2009b). La constitution de la distinction entre

intérieur et extérieur: proposition de recoupement entre Freud et les

neurosciences modernes. In: Neurosciences et psychothérapie:

Met opmaak: Frans (standaard)

Met opmaak: Frans (standaard)

Met opmaak: Nederlands (standaard)

28

Convergences ou divergences?, J. Monzée (red.) Ed. Liber, Montréal,

127-156

Bleuler, E. (1911). Dementia Praecox or the Group of Schizophrenias. New York,

International Universities Press.

Bonnard, (1967). Primary process phenomena in the case of a borderline

psychotic child. International Journal of Psychoanalysis, 48, 221-236.

Brakel, L. A. W. (2004). The psychoanalytic assumption of the primary process:

Extrapsychoanalytic evidence and findings. Journal of the American

Psychoanalytic Association, 52, 1131-112.

Brakel, L.A.W. (2009). Philosophy, Psychoanalysis and the A-Rational Mind.

Oxford: Oxford University Press.

Brakel, L. A. W., Kleinsorge, S., Snodgrass, M., & Shevrin, H. (2000). The

primary process and the unconscious: Experimental evidence supporting

two psychoanalytic presuppositions. International Journal of

Psychoanalysis, 81, 553–569.

Brakel, L. A. W. & Shevrin, H. (2005). Anxiety, attributinal thinking and the

primary process. International Journal of Psychoanalysis, 86, 1679-1693.

Brakel, L. A. W., Shevrin, H. & Villa, K.K. (2002). The priority of primary

process categorizing: experimental evidence supporting a psychoanalytic

developmental hypothesis. Journal of the American Psychoanalytic

Association, 50, 483-505.

Carhart-Harris, R. & Friston, K, (2010). The default mode, ego-functions and

free-energy: a neurobiological account of Freudian ideas. Brain, 133,

1265-1283.

Met opmaak: Engels(Groot-Brittannië)

Met opmaak: Engels(Groot-Brittannië)

29

Daniel, M.-P., Mores, C., Carite, L., Boyer, P., & Denis, M. (2006). Dysfunctions

of spatial cognition: the case of schizophrenic patients. Cognitive Process

7 (Suppl. 1):S173 DOI 10.1007/s10339-006-0131-1.

Ebert, J.N., Ewing, J.H., Rogers, M.H., & Reynolds, D.J. (1977). Changes in

primary process expression in hospitalized schizophrenics treated with

phenothiazines: Two projective tasks compared. Journal of Genetic

Psychology, 130, 83–94.

Fenichel, O. (1945). The psychoanalytic theory of neurosis. New York : Norton.

Fisher, C. (1954). Dreams and perception: the role of preconscious and primary

modes of perception in dream formation. Journal of the American

Psychoanalytic Association, 2, 389-445.

Fisher, C. (1957). A study of the preliminary stages of the construction of dreams

and images. Journal of the American Psychoanalytic Association, 5, 5-60.

Freud, S. (1953). The interpretation of dreams. In J. Strachey (Ed. & Trans.), The

standard edition of the complete psychological works of Sigmund Freud

(Vol. V, pp. 339-751). London: Hogarth Press. (Original work published

1900)

Freud, S. (1957). The Unconscious. In J. Strachey (Ed. & Trans.), The standard

edition of the complete psychological works of Sigmund (Vol. XIV, pp.

159-215). London: Hogarth Press. (Original work published 1915)

Freud, S. (1958). Formulations on the two principles of mental functioning. In J.

Strachey (Ed. & Trans.), The standard edition of the complete

psychological works of Sigmund Freud (Vol. 12, pp. 218-226). London:

Hogarth Press. (Original work published 1911)

30

Freud, S. (1960). Jokes and their relation to the unconscious. In J. Strachey (Ed. &

Trans.), The standard edition of the complete psychological works of

Sigmund (Vol. VIII). London: Hogarth Press. (Original work published

1905)

Freud, S. (1966). Project for a scientific psychology. In J. Strachey (Ed. &

Trans.), The standard edition of the complete psychological works of

Sigmund (Vol. I, pp. 281–392). London: Hogarth Press. (Original work

published 1950)

Friedman-Hill, S.R., Robertson, L.C., Desimone, R., & Ungerleider, L.G. (2003).

Posterior parietal cortex and the filtering of distractors. Proceedings of the

National Academy of Sciences of the USA, 7, 4263–4268.

Hamker, F. H. (2003). The reentry hypothesis: linking eye movements to visual

perception. Journal of Vision, 3, 808–816.

Holt, R. (1956). Gauging primary and secondary processes in Rorschach

responses. Journal of Projective Techniques, 22, 14-25.

Holt, R. (1967). The development of the primary process: a structural view. In R.

Holt (Ed.), Motives and Thought: Essays in Honor of David Rapaport (pp.

345-383). New York: International Universities Press.

Holt, R. (1977). A method for assessing primary process manifestations and their

control in Rorschach responses. In M. Rickers-Ovsiankina (Ed.),

Rorschach psychology (pp. 375-420). New York: Wiley.

Holt, R. (2002). Quantitative research on the primary process: Method and

findings. Journal of the American Psychoanalytic Association, 50, 457-

482.

31

Holt, R. (2007). Primary process scoring for the TAT. In S. R. Jenkins (Ed.) A

handbook of clinical scoring systems for thematic apperceptive techniques

(pp. 283-292). New York: Routledge.

Holt, R. (2009). Primary process thinking. Lanham: Jason Aroson.

Holt, R., & Havel, J. (1960). A method for assessing primary and secondary

process in the Rorschach. In M. Rickers-Ovsiankina (Ed.), Rorschach

psychology (pp. 283-315). New York: Wiley.

Kasanin, J. S. (1944). Language and thought in schizophrenia (Ed.). New York:

W. W. Norton.

Leys, C., & Schumann, S. (in press). A nonparametric method to analyze

interactions: The adjusted rank transform test, Journal of Experimental

Social Psychology.

Martindale, C., & Dailey, A. (1996). Creativity, primary process cognition and

personality. Personality and Individual Differences, 20(4), 409-414.

Nachmansohn, M. (1951). Concerning experimentally produced dreams. In D.

Rapaport (Ed.) Organization and Pathology of Thought (pp. 257–287).

New York: Columbia University Press. (Original work published 1925)

Pomarol-Clotet, E., Salvador, R., Sarró, S., Gomar, J., Vila, F., Martínez, A., et al.

(2008). Failure to deactivate in the prefrontal cortex in schizophrenia:

dysfunction of the default mode network? Psychological Medecine, 38,

1185–1193.

Pötzl, O. (1917). The relationship between experimentally induced dream images

and indirect vision. In Psychological Issues Monograph 7 (1960)

Met opmaak: Engels(Groot-Brittannië)

32

Preconscious Stimulation in Dreams, Associations, and Images – Classical

Studies, volume 2, pp. 41-120.

Raichle, M.E., MacLeod, A.M., Snyder, A.Z., Powers, W.J., Gusnard, D.A.,

Shulman, G.L. (2001). A default mode of brain function. Proceedings of

the National Academy of Sciences USA, 98, 676–682.

Rapaport, D. (1951). States of consciousness. In M. Gill (Ed.), The Collected

Papers of David Rapaport (pp. 385-404). New York: Basic Books.

Rapaport, D., Gill, M.M., & Schafer, R. (1945-46). In R.R. Holt (Ed.) Diagnostic

Psychological Testing. Chicago: Yearbook Publishers; revised edition.

New York: International Universities Press.

Robbins, M. (2002). The language of schizophrenia and the world of delusion.

International Journal of Psychoanalysis, 83(2), 383-405.

Roffenstein, G. (1951). Experiments on symbolization in dreams. In D. Rapaport

(Ed.) Organization and Pathology of Thought (pp. 249–256). New York:

Columbia University Press. (Original work published 1924)

Roulot, D. (2004). Schizophrénie et Langage: que veut dire le mot chapeau?

Toulouse: Eres.

Rousselet, G.A., Thorpe, S.J., & Fabre-Thorpe, M. (2004). How parallel is visual

processing in the ventral pathway? Trends in Cognitive Sciences, 8, 363–

370.

Saretsky, T. (1966). Effects of chlorpromazine on primary-process thought

manifestations. Journal of Abnormal Psychology, 71(4), 247-252.

33

Schrötter, K. (1951). Experimental dreams. In D. Rapaport (Ed.) Organization

and Pathology of Thought (pp. 234–248). New York: Columbia

University Press. (Original work published 1911)

Shevrin, H. (1973). Brain wave correlates of subliminal stimulation, unconscious

attention, primary and secondary process thinking, and repression. In M.

Mayman (Ed.) Psychoanalytic Research: Three Approaches to the Study

of Subliminal Processes (pp.56–87). Psychological Issues, 30. New York:

International Universities Press.

Shevrin, H., & Luborsky, L. (1961). The rebus technique: a method for studying

primary process transformations of briefly exposed pictures. Journal of

Nervous and Mental Disease, 133(6), 479-488.

Spielberger, C.D. (1979). Preliminary manual for the State-Trait Personality

Inventory. Tampa: U Florida Press.

Spielberger, C.D., Gorsuch, R.L., & Lushene, R.E. (1970). Manuel for the State-

Trait Anxiety Inventory (Self-evaluation Questionnaire). Palo Alto,

California: Consulting Psychologists Press.

Spielberger, C.D., Gorsuch, R.L., Lushene, R.E., Vagg P.R., & Jacobs, G.A.

(1993). Manuel de l‟Inventaire d‟anxiété état-trait forme Y (STAI-Y).

Adapted to French by M. Bruchon-Schweitzer et I. Paulhan. Paris:

Editions du Centre de Psychologie Appliquée.

Van de Vijver, G., Bazan, A., Rottiers, F., & Gilbert J. (2006) Enactivisme et

internalisme: de l'ontologie à la clinique. Intellectica 43/1, 93-103.

Van Der Ploeg, H.M., Defares, P.B., & Spielberger, C.D. (1980). Handleiding bij

de Zelf-Beoordelings Vragenlijst, ZB.: Een Nederlandstalige bewerking

34

van de Spielberger State-Trait Anxiety Inventory. Lisse: Swets &

Zeitlinger.

Vanheule, S., Roelstraete, B., Geerardyn, F., Murphy, C., Bazan, A., Brakel,

L.A.W. (2011). Construct validation and internal consistency of the

geometric categorization task (GEOCAT) for measuring primary and

secondary processes. Psychoanalytic Psychology, 28, 2, 209-228.

Villa, K. K., Shevrin, H., Snodgrass, M., Bazan, A., Brakel, L. A. W. (2006).

Testing Freud‟s hypothesis that word forms and word meanings are

functionally distinct in the unconscious: Subliminal primary process

cognition and its links to personality. Neuro-Psychoanalysis, 8, 117-137.

von Domarus, E. (1944). The specific laws of logic in schizophrenia. In J. S.

Kasanin (Ed.), Language and Thought in Schizophrenia (pp. 104–114).

New York: Norton.

Whitfield-Gabrieli, S., Thermenos, H.W., Milanovic, S., Tsuang, M.T., Faraone,

S.V., McCarley, R.W., et al. (2009). Hyperactivity and hyperconnectivity

of the default network in schizophrenia and in first-degree relatives of

persons with schizophrenia. Proceedings of the National Academy of

Sciences USA, 106, 1279–84.

35

ACKNOWLEDGEMENTS

The authors wish to thank all the patients in the three different psychiatric

institutions (LaBorde, Psychiatric Centres Sint-Amandus and Guislain) for their

participation. Acknowledgements also go to Stijn Van Eeckhoven, Bram

Herrebout, Bjorn Roelstraete and Christophe Leys for help with data-gathering

and analysis. Finally, we wish to thank Gertrudis Van de Vijver and Dr. Georges

Otte for useful comments and encouragements. This research was supported by

the Belgian American Educational Foundation, Howard Shevrin and the Society

for Neuro-Psychoanalysis.

36

Figure 1: Distributions of attributional responses in the psychiatric population of

the present study (left) and in a healthy (non-psychiatric) population of

comparable age (LifeCat study, Brakel et al., 2002; right).

0 1 2 3 4 5 6 Number of attributional responses

0

10

20

30

40

50

PSYCHIATRIC PATIENTS

Maximum = 6

AcuteCat (18-68 yrs)

Nu

mb

er

of

pa

rtic

ipa

nts

0 1 2 3 4 5 6 Number of attributional responses

0

20

40

60

80

100

120

140

Maximum = 6

NORMAL POPULATION

LifeCat (19-69 yrs)

37

Figure 2: Distributions of the attributional responses in the psychotic patients

(left) and in the non-psychotic psychiatric patients (right).

38

Figure 3: Distributions of the attributional responses in the patients with (left) and

without (right) acute psychotic symptoms.

39

Table 1: Mean (± SEM), median and mode number of attributional

responses (out of maximum 6) and percentage of participants favouring

attributional (ATT) above relational (REL) choices in a psychiatric population

(AcuteCat, present study) versus in a non-psychiatric population of comparable

age (LifeCat, Brakel et al., 2002); mean anxiety score (± SEM) on the Spielberger

State Trait Anxiety Inventory (STAI).

Study population (age) n mean median mode % ATT > REL STAI

AcuteCat total (18-68) 127 3.66 ± 0.21* 4 6 59.8* 45.0 ± 1.2**

LifeCat adults (19-69) 254 1.76 ± 0.14 0 0 24.0 /

* p < 0.001 as compared to non-psychiatric participants of comparable age; one-

tailed

** p<.001 as compared to the Dutch (36.4) or French (35.73) norms for a

population of comparable age

40

Table 2: Mean (± SEM), median and mode number of attributional responses (out

of maximum 6) and percentage of participants favouring attributional (ATT)

above relational (REL) choices in psychotic patients versus non-psychotic

patients; mean anxiety score (± SEM) on the STAI.

Patient population n mean median mode % ATT > REL STAI

Psychotic 72 3.97±.27* 5 6 65.3 44.7 ± 1.5

Non-Psychotic 55 3.25±.34 4 6 52.8 45.4 ± 2.0

* p =.060 as compared to non-psychotic patients; one-tailed

41

Table 3: Mean (± SEM), median and mode number of attributional responses (out

of maximum 6) and percentage of participants favouring attributional (ATT)

above relational (REL) choices in patients with and without acute psychotic

symptoms (); mean anxiety score (± SEM) on the STAI.

Patient population n mean median mode % ATT > REL STAI

with acute

psychotic

36 4.31±.37* 6 6 65.3 42.4 ± 2.3

without 91 3.41±.26 4 6 52.8 46.0 ± 1.5

* p <.05 as compared to patients without acute psychotic symptoms; one-tailed

42

i Italics added

ii The GeoCat forms were administered with the help of the “naïve” researchers

Stijn Van Eeckhoven and Bram Herrebout, who were blind to the hypotheses and

to the test rationale.

iii For use of the GeoCat, please contact Pr. Brakel (brakel@umich.edu).

iv Medians are a useful measure especially for skewed distributions.

v This leads to uneven groups for comparison, with n = 36 for the patients with

acute psychotic symptoms and n = 91 (36 psychotic patients + 55 non-psychotic

patients) for the patients without acute psychotic symptoms. However, as we are

using non-parametric statistics, which do not require the assumption of

homogeneity of variance, this is not a problem.